In addition to harm reduction approaches, the combined process of drug- free treatment and recovery is a significant HIV risk reduction strategy for woman who inject drugs and engage in risky sexual activities. The premise of the proposed studies that drug-free recovery directly reduces injection drug use and crack use, and indirectly reduces sexual risk behaviors. Further, social support--defined as "resources provided by others"--is essential to maintaining drug-free recovery and influences the adoption and constant practice of HIV risk reduction behaviors over time. The aims of the study are: (1) To document longitudinally the sources, types, amount, and quality of social support for women ex- offenders during the first seven months after leaving drug treatment; (2) to assess ethnographically the function of various types of social support in relation to HIV risk reduction behaviors; (3) to examine the interrelationships between various behavioral indicators of HIV/AIDS risk reduction and measures of HIV risk behaviors, social support, drug treatment, and individual characteristics (including HIV serostatus, psychological traits, and demographics) as well as variations in these relationships over time; (4) to develop models to predict HIV related risk reduction behaviors among women ex-offenders who have been mandated to drug synthesize research findings and recommend HIV intervention strategies. To achieve the aims, the proposed study will conduct baseline interviews with women (N-600) recently admitted to drug treatment in prison/jail and community based programs and follow them up 30-days and 7-months after they leave treatment or prison/jail. The study integrates qualitative and quantitative data and methods. A semistructured questionnaire will contain open-ended questions to obtain qualitative information on the sources and types of social support, and close-ended questions to obtain quantitative data on HIV related behaviors and individual characteristics. Several standard psychological instruments will also be administered. An ethnographic study of a subsample of treatment graduates (N=60) will also be conducted. In an area barren with research, this study will provide insight about how social support influences HIV risk reduction behaviors among women in recovery.